北京测绘2025，Vol.39Issue(1)：1-7,7.DOI:10.19580/j.cnki.1007-3000.2025.01.001
基于深度学习的地理要素类别语义匹配研究进展
Research progress on deep learning-based semantic matching of geographical element categories
摘要
Abstract
Geospatial data resources are becoming increasingly abundant as geographical data collection methods advance.However,semantic heterogeneity in geospatial data under different classification systems prevents effective data fusion in applications,resulting in the problem of"supply and demand dislocation of geospatial data".Semantic matching is the key to solving this problem,but current methods mainly rely on expert knowledge and have limited scalability.This paper discussed the progress of semantic matching of geographical element categories,with a focus on similarity calculation and vectorization representation methods.It also explored the semantic matching feasibility of geographical element categories in the absence of expert knowledge.On this basis,the research opportunities for semantic matching of geographical element categories were proposed.关键词
非监督学习/语义匹配/地理要素类别/深度神经网络Key words
unsupervised learning/semantic matching/geographical element category/deep neural network分类
天文与地球科学引用本文复制引用
蔡荣锋,谭永滨,王宏..基于深度学习的地理要素类别语义匹配研究进展[J].北京测绘,2025,39(1):1-7,7.基金项目
国家自然科学基金(42361067) （42361067）
东华理工大学2024年度研究生创新专项资金(DHYC-202411). （DHYC-202411）
